Extensive genome-wide single nucleotide polymorphisms (SNPs) are now available. Theoretically, we can systematically consider all the regions of the genome to identify those regions associated with disease susceptibility or unfavorable risk factors. Important issues need to be resolved however, before we can practically use all the genetic information in genome-wide association studies. Methods need to be developed to detect and resolve genotyping errors and we need new, analytical strategies designed specifically for family data that will account for multiple comparisons. [unreadable] [unreadable] Many large family-based studies, including our own, were initiated with the goal of first detecting linkage to identify chromosomal regions likely to harbor mutations having relatively large effects on important risk factors for heart disease. One of our studies, the Heritability and Phenotype Intervention (HAPI) Heart Study, includes extensive coronary heart disease risk factor data and 500,000 SNPs on 900 adults in Old Order Amish pedigrees. Our families are very suitable for genome-wide association studies and they offer special opportunities, compared to population-based samples, because families provide a direct test of allelic transmissions. [unreadable] [unreadable] This application addresses two specific issues pertaining to genome-wide association studies in families. One relates to development, implementation, testing, and dissemination of efficient ways to clean and process genome-wide SNP data and then create haplotypes. The other relates to developing analytic strategies that combine information from population- and transmission-based association tests to improve power, minimize false positive rates, and enhance efficiency for detecting SNP-trait associations. [unreadable] [unreadable] [unreadable] [unreadable]